J . Am. Chem. SOC.1985, 107, 5204-5209
5204
Kinetics and Mechanism of the Hydrolysis of Acyclic Oxyphosphonium Salts Robert A. McClelland,* Glenn H. McGall, and Geeta Patel Contribution from the Department of Chemistry, University of Toronto, Toronto, Ontario, Canada M5S 1A l . Received December 17. 1984
Abstract: A kinetic study is reported of the hydrolysis of seven acyclic alkoxyphosphonium ions [4-X-C6H,( Me)P(OMe)2+ (X = H, CI, Me), PhP(OR)3+ (R = Me, Et), Ph2P(OMe)*+,Ph2(Me)P+OMe] and two alkoxyaryloxyphosphonium ions [4-XC6H4(Me)P(OMe)OC6H4-4-OMe+ (X = H, OMe)]. First-order rate constants follow the relationship koM = ko + koHIOH-] + k,[A-], showing a reaction with water, hydroxide, and the base component of the buffer. The last two reactions are shown or argued to involve only P-0 cleavage in the hydrolysis. The water reaction occurs with some C-0 cleavage as well. The mixed phosphonium ions hydrolyze over a range of acidities to produce 4-methoxyphenol and the methyl phosphinate ester. The P-0 cleavage reaction in all cases is argued to involve rate-limiting formation of a pentavalent hydroxyphosphorane intermediate. The buffer reaction represents general base catalysis of the addition of water, the base assisting the water attack by simultaneously removing a proton. Bransted coefficients are large, ranging from 0.6 to 0.85. This implies a transition state with a considerable amount of proton transfer. A comparison is presented with the general base-catalyzed hydration reactions of carbocations, which show significantly smaller 0values. The water rates for the phosphonium ions lie close to the Bransted plots, suggesting that this reaction is also occurring with general base catalysis. Large rate decreases are observed with added acids and salts. These effects are explained by the general base mechanism with considerable hydronium ion character in the transition state. A comparison is also presented of the effects of sulfuric acid on the rates of hydrolysis of an alkoxyphosphonium ion and an analogous hydroxyphosphonium ion or protonated phosphinate ester.
The hydrolysis of phosphate esters has seen considerable in part because of the importance of such species in biological systems. T h e acid-catalyzed hydrolysis involves The equilibrium protonation followed by reaction with
? -P-OR
I
I
H‘
-
L
OH -P’-OR
HP
I
on -;LOR
OC H,
Ph 2
k,
-PbAr
+
OPh Ph-btOPh
CH,P(OPh),
CH,
3
+
I
Prod
protonated ester 1, a hydroxyphosphonium ion, is modeled by alkoxy- and aryloxyphosphonium ions where the labile hydrogen is replaced by an alkyl or aryl group. The hydrolysis of such ions is therefore of interest as a model of the slow step in the ester hydrolysis. Previous studies with alkoxyphosphonium salts have generally been carried out in highly basic solutions and have focussed on the stereochemistry of the substitution reaction of phosphorus.6 The reactive strained phosphetanium ion 2 has been subjected to a detailed study over a range of p H in 50% acetonitri1e:water.’
pf-
to be highly reactive, necessitating the use of mixed aqueous solvents to slow the hydrolysis to allow kinetic study. One prominent feature of these studies was the observation of pronounced rate decreases on the addition of salts or acids. The following mechanism was proposedg to account for this. A change
4
A complex rate-pH profile was obtained, being interpreted in terms of a two-step reaction with a pentavalent hydroxyphosphorane intermediate, with a change in rate-limiting step with changing pH. Aryloxyphosphonium ions (for example, 3 and 4) have also been i n ~ e s t i g a t e d ,although ~,~ these cations were found (1) Westheimer, F. H . Acc. Chem. Res. 1968, 1 , 70-78. (2) Bunton, C. A. Acc. Chem. Res. 1970, 3, 257-265. (3) Hudson, R. F.; Brown, C. Acc. Chem. Res. 1972, 5, 204-211. (4) Cook, R. D.; Abbas, K. A. Tetrahedron Lett. 1973, 519-520. Abbas, K. A,; Cook, R. D. Tetrahedron Lett. 1975, 3559-3562. (5) Bunnett, J. F.; Edwards, J. 0.;Wells, D. V.; Brass, H. J.; Curci, R. J . Org. Chem. 1973, 38, 2703-2707. (6) (a) p e Bruin, K. E.; Zon, G.; Naumann, K.; Mislow, K. J . Am. Chem. SOC.1969, 91, 7023-7027. (b) De Bruin, K. E.; Zon, G.; Naumann, K.; Mislow, K. Ibid. 1969, 91, 7027-7030. (c) De Bruin, K. E.; Mislow, K. Ibid. 1969, 91, 7393-7397. (d) De’Ath, N. J.; Trippett, S. J . Chem. SOC.,Chem. Commun. 1969, 172-173. (e) Lewis, R. A,; Naumann, K.: De Bruin, K. E.; Mislow, K. Ibid. 1969, 1010-1011. ( f ) De Bruin, K. E.; Jacobs, M. J. Ibid. 1971, 59-60. (g) Reiff, L. P.; Szafraniec, L. J.; Aaron, H . S . Ibid. 1971, 366-367. (h) De Bruin, K. E.; Petersen, J. R. J . Org. Chem. 1972, 37, 2272-2278. (i) Mislow, K. Acc. Chem. Res. 1970, 3, 321-331. (7) Gorenstein, D. G. J . Am. Chem. Soc. 1973, 95, 806G8065.
0002-7863/85/1507-5204$01.50/0
HbH,
H, I
-:7
3 H , O e h-1
OAr
Y no
R
-6.-
-p-
I
I
OA r
in rate-limiting step with changing acidity was also invoked here, the water addition being rate limiting a t low acid concentrations with the second step becoming rate limiting a t high acid concentrations. The changeover was proposed to occur because of the requirement for acid in the reverse of the addition (the kwl step). In our studies on the hydrolysis of alkoxyphosphoranes’@’* (for example 5 ) , we have observed an acid-catalyzed reaction whose mechanism is directly analogous to that normally associated with the hydrolysis of acetals and ortho ester^.^^^'^ An alkoxy-
7’2
Ph-Pc0
O A O H
n’ I + Ph-P*.o
L J
b
5
6
HsO
+Prod
(3)
(8) Lonzetta, C. M.; Kubisen, S. J.; Westheimer, F. H . J . Am. Chem. SOC. 1976, 98, 1632-1634. (9) Kubisen, S. J.; Westheimer, F. H. J . Am. Chem. SOC. 1979, 101, 5985-5990. (10) McClelland, R. A.; Patel, G.; Cirinna, C. J . Am. Chem. SOC.1981, 103, 6432-6436. (1 1) McGall, G.; McClelland, R. A. J. Chem. SOC.,Chem. Commun. 1982, 1222-1 223. (12) For other kinetic studies of phosphorane hydrolysis see: (a) Archie, W. C.; Westheimer, F. H. J . Am. Chem. SOC.1973, 95, 5955-5959. (b) Queen, A,; Lemire, A. E.: Janzen, A. F. I n t . J . Chem. Kinef. 1981, 13, 411-416. (c) Belskii, V. E.; Khismatullina, L. A,; Bykova, T. G.; Burykina, A. V.; Ivanov, B. E. J . Gen. Chem. USSR 1979, 49, 298-302.
0 1985 American Chemical Society
J. Am. Chem. SOC., Vol. 107, No. 18, 1985 5205
Hydrolysis of Acyclic Oxyphosphonium Salts phosphonium ion is an intermediate in this reaction, and our studies have revealed quite pronounced differences in the hydrolytic lability of cyclic and acyclic systems, paralleling those observed in the hydrolysis of cyclic and acyclic phosphate esters.] The cyclic phosphonium ion 6 , for example, is not observed during the hydrolysis of 5 and was estimated to have a half-life in water of less than 1 ms.Io On the other hand, the acyclic ions PhP+(OR)3 derived from the phosphoranes PhP(OR)4 have half-lives of the order of minutes or even hours.15 In order to obtain a more thorough understanding of phosphonium ion chemistry, we have carried out a study of the hydrolysis as a function of acidity, salt concentration, and buffer concentration of a series of alkoxyphosphonium ions 7-13 and two alkoxyaryloxyphosphonium ions 14 and 15. These cations
Phosphonium ions 7-10, 12, and 14 were prepared by dissolving the appropriate phosphinate or phosphonate ester in dry CH2CI2(distilled from concentrated H,SO,) to give a solution 0.1-1 .O molar in concentration. Methyl trifluoromethanesulfonate (MeOTf (Aldrich)) was added and the solution allowed to stand a t ambient temperature. The progress of the methylation was monitored by recording ' H N M R and 31P N M R spectra of the ~ o l u t i o n . ~ ' As an example, methyl methylphenylphosphinate exhibits a l H N M R spectrum with its P-methyl doublet at 2.02 ppm and its 0-methyl doublet at 3.80 ppm. Addition of MeOTf results in a decrease in intensity of these peaks, with new doublets now in a 1:2 ratio appearing a t 2.46 and 4.16 ppm. A corresponding change is seen in the 31P N M R with the phosphinate ester signal at 29.4 ppm giving way to the phosphonium ion signal at 78.4 ppm. Lewis and ColleZ8have reported 31P chemical shift of 78.5 ppm for this ion. The reactions achieve their equilibrium position within a day; as reported by Lewis and Colle28 the methylations d o not proceed completely to the phosphonium ion. Approximate values of the equilibrium constant for X3+0 MeOTf X3P+-OMe.0Tf were determined from relative N M R peak intensities and are 5 for the cations 7, 8, 9, and 12, 1 for the cation 10, and 0.5 for the cations 14 and 15. In the last three cases an excess, sometimes as much as 10-fold, of MeOTf was employed in order to achieve an excess of phosphonium ion in solution. Phosphonium ions 7 and 12 were also prepared by reacting in CH2Clzthe phosphinate ester and an equivalent amount of trimethyloxonium tetrafluoroborate. In this case the phosphonium ion as its BF,- salt could be precipitated by the addition of excess ether (distilled from sodium). These salts, however, proved highly hygroscopic and tended to turn to an oil even on standing under strictly anhydrous conditions. The phosphonium ion 13 was prepared by the P-methylation in CH2C1, of methyl diphenylphosphinite (Aldrich) with MeOTf as described by Colle and Lewis.29 The phosphonium ion 11 was prepared in two ways, by ethylation of diethyl phenylphosphonate with triethyloxonium tetrafluoroborate and by the addition of tetraethoxyphenylphosphorane directly to aqueous solutions. This phosphorane produces the ion extremely rapidly when added to water.Is Kinetics. Instruments employed for kinetic studies were Unicam sp 1800 and Cary 2390 spectrometers and a Durrum Gibson stopped-flow spectrometer. Conventional UV kinetic studies were carried out by injecting 1-3 p L of the CH2CI2solution containing the phosphonium ion or phosphorane directly into the thermostated UV cuvette containing 2-3 mL of aqueous solution. Final concentrations of phosphonium ion were of the order 5-10 X molar. The wavelengths chosen for kinetic studies were in the range 275-280 nm where a significant decrease in absoLbance accompanies hydrolysis (see Results and Discussion). First-order rate constants were determined as the slopes of plots of In ( A - A , ) vs.time. In those experiments conducted with cations with the CF3S03-counterion and with 11 when generated from the phosphorane, excellent linearity over 4-5 half-lives was observed and excellent reproducibility (*2%) in duplicate runs was obtained. Considerable problems were, however, encountered when BF4- was employed as the counterion, both with respect to the linearity of individual runs and to the reproducibility. The source of these problems could not be discovered although we did note that the addition of NaBF, markedly accelerates the hydrolysis. In any event kinetic experiments using the BF4- counterion were abandoned. As described in the materials section an equilibrium mixture of phosphonium ion, ester, and MeOTf was usually present in the CH2CI, solutions. Kinetic experiments were conducted with the cation 10 prepared with differing relative amounts of the alkylating agent ( l : l , 4:1, and 20:l with respect to ester). N o change outside of experimental error was observed in the rate constant, showing that the presence of unreacted ester or MeOTf does not affect the kinetics. The latter species is hydrolyzed as well,'O but since there is no chromophore involved the kinetic studies in question are not affected (providing sufficient buffer is present to neutralize the acid which is formed from the MeOTf hydrolysis). Kinetic studies in the more basic solutions were carried out on the stopped-flow spectrophotometer by adding the phosphonium ion to 0.002 M HCI solution and mixing this solution on the stopped-flow apparatus with the appropriate basic solution. Products. Products of the hydrolysis of the cation 7 were determined by dissolving the isolated BF,- salt in D,O to give solutions of about 1 molar and recording directly the 'H NMR spectral changes. The position of bond cleavage in the hydrolysis of this cation was determined by adding the isolated BF, salt prepared by methylating the methoxy-180 phosphinate with Me30+.BF4- to water and to an 0.2 molar acetate
+
1,X:H
lO.R:CH, 1 1 , R : CH,CH,
8,x:cI
12
9,X:CH
OCH, P h-
6I.-C H, Ph
13
X
e
F : e C H , CH,
14. X : H
15, x :ocn,
are all of sufficient stability to permit their kinetic study in wholly aqueous solutions. Perhaps the most notable feature is a very pronounced catalysis of the hydrolysis by added buffers. A comparison is also carried out with the hydrolysis or hydration of oxycarbonium ions and other stabilized carbocations.'6-20 Experimental Section Materials. Methyl methylarylphosphinates (aryl phenyl, 4-chlorophenyl, 4-methylphenyl) were prepared as described by Edwards and c o - w ~ r k e r s .(Methoxy-I80)methyl ~ methylphenylphosphinate containing 1 1.3% excess was preparedS from the phosphinic acid chloridez3and [1gO]-methanol.24 Methyl diphenylphosphinate, dimethyl phenylphosphonate, and diethyl phenylphosphonate were prepared by treatment of diphenylphosphinic acid chloride (Aldrich) and phenylphosphonic acid chloride (Aldrich) with the corresponding alcohol, following a published procedure.2s 4-Methoxyphenyl methylarylphosphinates with aryl phenyl and 4-methylphenyl were prepared in a similar manner by the reaction of the phenol and acid chloride. Tetraethoxyphenylphosphorane was prepared by using the sulfenate ester route of Denney and coworkers,26 reacting ethyl phenylsulfenate with diethyl phenylphosphonate. (13) Cordes, E. H.; Bull, H. G.Chem. Reu. 1974, 74, 581-603. (14) Ahmad, M.; Bergstrom, R. G.; Cashen, M. J.; Chiang, Y.; Kresge, A. J.; McClalland, R. A,; Powell, M. J. J . A m . Chem. SOC.1979, 101, 2669-2677. (15) McGall, G.; McClelland, R. A,, work in progress. The acyclic phosphoranes PhP(OR), are highly reactive toward cleavage producing PhP+(OR)>,but the kinetics of this process can be studied in organic solvents containing small amounts of water. (16) Young, P. R.; Jencks, W. P. J . Am. Chem.Soc. 1977,99,8238-8248. ( 1 7 ) McClelland, R. A,; Ahmad, M. J . Am. Chem. SOC.1978, 100, 7027-703 1. (18) McClelland, R. A,; Ahmad, M . J . Am. Chem. SOC.1978, 100, 7031-7036. (19) Okuyama, T.; Fujiwara, W.; Fueno, T. J. Am. Chem. SOC.1984, 106, 657-662. (20) (a) Ritchie, C. D. Acc. Chem. Res. 1972, 5, 348-354. (b) Ritchie, C. D. J . Am. Chem. SOC.1972, 94, 3275-3276. (c) Ritchie, C. D. Ibid. 1975, 97, 1170-1197. (d) Ritchie, C. D.; Wright, D. J.; Huang, D.S.; Kamego, A. A. Ibid. 1975, 97, 1163-1 170. (21) Ride, P. H.; Wyatt, P. A. H.; Zochowski, Z. W. J . Chem. Soc., Perkin Trans. 2, 1974, 1188-1 189. (22) Bunton, C. A,; Davoudazedeh, F.; Watts, W. E. J . Am. Chem. SOC. 1981, 103, 3855-3858. (23) Korpium, 0.;Lewis, R. A,; Chickos, J.; Mislow, K. J . Am. Chem. SOC.1968, 90, 4842-4846. (24) Sawyer, C. B. J . Org. Chem. 1972, 37, 4225-4226. (25) Edmunson, R. S.;Wrigley, J. 0. L. Tetrahedron 1967, 23, 283-290. (26) Chang, L. L.; Denney, D. B.; Denney, D. Z.; Kazior, R. L. J . Am. Chem. SOC.1977, 99, 2293-2297.
-
(27) The peak in the IH NMR due to the CH2C12solvent, a singlet near 5.3 ppm, does not interfere. (28) Lewis, E. S.; Colle, K. S . J . Org. Chem. 1981, 46, 4369-4372. (29) Colle, K. S.; Lewis, E. S . J . Ora. Chem. 1978, 43, 571-574. (30) McGarrity, J. F.; Prodolliet, J. W.Tetrahedron Lett. 1982, 417-420. ~~
5206
J . Am. Chem. Soc., Vol. 107, No. 18, 1985
McClelland, McGall, and Patel
buffer. After a time corresponding to 10 half-lives of hydrolysis, the phosphinate ester was isolated by extraction with CH2C12and analyzed directly on a mass spectrometer by measuring the ratio 172:170 corresponding to the molecular ions. The procedure employed here is the same as that used p r e v i ~ u s l y , and ~ ~ . the ~ ~ excess I80present in the ester was calculated by using the formula 99.8 (R - Ro)/( 1.002 + R - R,)where R is the 172:170 ratio of the sample in question and Rois the same ratio in a sample of unlabeled material. This equation corrects peak intensities for natural isotopic variation throughout the molecule including that associated with oxygen (0.2% natural abundance). Products of the hydrolysis of the cation 15 were determined as follows. A CH2C12solution of the phosphonium ion was prepared as previously described by methylation of the p-methoxyphenyl ester with a 20-fold excess of MeOTf. This solution contains 90% phosphonium ion and 10% unreacted ester. With a mechanical stirrer this solution was shaken with the appropriate aqueous solutions. After a time corresponding to 10-20 half-lives of hydrolysis the CH2C12was separated and dried (MgSOJ and the CH2C12removed on a rotary evaporator. The IH NMR of the remaining material was recorded in CDCI,, the ratio of methyl methyl(4-methylpheny1)phosphinate and 4-methoxyphenyl methyl(4-methylpheny1)phosphinate being determined by integration of the P-Me doublets near 2 ppm. Three aqueous solutions were employed, 35% H2S0,, 0.01 M HCI, and 0.2 M acetate buffer of pH 4.5, and in each case the ratio methyl esterpmethoxyphenyl ester was 1O:l. Results Products. Products of the hydrolysis of the phosphonium ion 7 were determined by monitoring the course of the reaction in an NMR spectrophotometer. Hydrolysis in both D20alone and in a concentrated acetate buffer proceeds cleanly to produce methyl methylphenylphosphinate and 1 equiv of methanol. UV spectral changes with the other alkoxyphosphonium ions are very similar to that observed with 7 (see later). W e assume that similar products arise with these, namely the alcohol plus phosphoryl compound - a phosphonate ester from 10 and 11, a phosphinate ester from 7, 8, 9, and 12, and a phosphine oxide from 13. The question of the position of the bond cleavage2%)) was also approached by hydrolyzing a n '*O-labeled version of the cation 7 prepared by methylating a methoxy-labeled ester. The resulting cation has two methoxy groups, so that four hydrolyses can occur,
(4)
CH
CH,
two P-0 cleavage reactions and two C-O cleavage reactions. As shown in eq 4, C-0 cleavage must produce a phosphinate which retains the label. P-O cleavage on the other hand results in labeled ester and unlabeled ester in equal amounts (ignoring isotope effects), so that overall 50% of the original label would be lost. The starting phosphinate ester contained 11.8 f 0.05% excess I8O (average of four measurements). Hydrolysis of the phosphonium ion in water alone results in a phosphinate with 6.7 f 0.1% excess l 8 0 (average of two experiments), implying that the hydrolysis occurs with 87 f 2% P-0 cleavage. This result is interesting in that the acid-catalyzed hydrolysis of the phosphinate ester itself occurs with 91% P-0 cleavage.)) The p a r a l l e l i ~ mof~the ~ cation and ester results suggests that the phosphinium ion does indeed provide a reasonable model for the protonated ester. The labeled cation was also hydrolyzed in 0.2 M acetate, p H 4.6, a solution where the kinetics show that a n acetate-catalyzed hydrolysis is the dominant reaction. In this experiment the ester (31) McClelland, R. A. J . Am. Chem. SOC.1975, 97, 3177-3181. (32) McClelland, R. A,; Ahmad, M. J . Am. Chem. SOC.1977, 99, 5356-5360. (33) McClelland, R. A. J . Chem. SOC.,Chem. Commun. 1976,226-227. (34) A difference in temperature does exist-cation (25 "C), ester (69 'C)-so that an exact comparison under identical conditions is not possible.
0.5
c
abs
Figure 1. UV spectral changes occurring during the hydrolysis of the phenylmethyldimethoxyphosphonium ion (7) in 0.01 M HCI. Spectra molar cation initially. were measured for solutions containing 5 X The time interval between each spectra is 10 min.
product was found to contain 5.5 f 0.2% excess l 8 0 (average of two experiments), implying, within experimental error, only P-0 cleavage. This result is also consistent with the observation of methanol as the product, and not methyl acetate in this solution. In general we conclude that similar patterns occur for the other catalysts and with the other phosphonium ions. The water reaction involves a mixture of P-0 and C-0 bond cleavages with a significant if not predominant amount of the former. The buffercatalyzed hydrolyses occur only with P-O cleavage. The similarity in the B r ~ n s t e dplots for each of the phosphonium ions (see later) is the principal argument that this common mechanistic behavior does exist. It can also be noted that several studies have been conducted with hydroxide as the nucleophile and only P-0 cleavage found.6 With the phosphonium ions 14 and 15 hydrolysis can occur in two ways, producing an aryl ester and methanol or a methyl ester and p-methoxyphenol. Products were determined from the phosphonium ion 15 in three solutions-35% H2S04,0.01 M HCI, and an acetate buffer of p H 4.6. The method of preparation of this phosphonium ion involves methylation of the aryl ester. Since these methylations do not proceed to completion (see Experimental Section), unreacted ester is present and thus appears in the mixture after hydrolysis. A mixture of 90% methyl ester and 10% aryl ester is observed after hydrolysis in each solution. The phosphonium ion solution, however, contained 10% aryl ester, implying that this is the source of this compound in the hydrolysis product. W e conclude therefore that the cation itself hydrolyzes with loss of the phenol. Kinetics. Figure 1 shows a typical absorbance change for the hydrolysis. The spectra of the phosphonium ions and ester products are similar, but the cations in every case are more strongly absorbing and show a slight shift to longer wavelength. Quite significant absorbance changes are therefore observed and kinetic measurements performed routinely. All of the phosphonium ions of this study are of sufficient stability in water that their hydrolysis can be conveniently studied by conventional U V spectroscopy. The disappearance of the phosphonium ions was studied a t constant p H in HCI solutions or buffer solutions. Excellent adherence to the first-order rate law was observed. Rate constants determined in HC1 or by extrapolation to zero buffer concentration obeyed the simple relationship
J . Am. Chem. Soc., Vol. 107, No. 18, 1985 5201
Hydrolysis of Acyclic Oxyphosphonium Salts
Table I. Rate Constants for the Hydrolysis of the Phenylmethyldimethoxyphosphonium Ion (7) (25 OC, ionic strength ( b ) = 0.5) and the Phenyltriethoxyphosphonium Ion (11) (25 OC, = 0.1)
k (M-' H2O
D2O CNCH2C02ClCHzCO2CH30CH2CO; CICH2CH2COZCH3C02CH,CO,-(in D 2 0 ) F (CH3)2As02-
HP0,-2 N-methylmorpholine
11
0.00013 s-I 0.000038 s-I 0.001 1 0.0017 0.0067 0.015 0.063 0.044 0.02 0.94 2.48 2.32
tris
.2
1
.3
rA-1
Figure 2. First-order rate constants, s-l, for the hydrolysis of the phenylmethyldimethoxyphosphonium ion (7) as a function of the molar carboxylate concentration. The point X is the average of rate constants measured in HCI solutions.
0.00024
s-l
0.001 1 0.0034 0.022
0.16 0.19 0.14 0.6
36 83 5.3
0
s-I)
7
catalyst
5.6 4.9 x 102
x 104
alcohol (or phenol) as in the k2/ reaction, or at the very least, this group can provide a hydrogen bond to solvent water molecules. In the latter case the forward and reverse reactions each result in phosphonium ions, but the argument is that the hydroxyphosphonium ion is more stable because of the solvent interaction. Some evidence that this is the case will be presented later. There is some evidence pointing to rate-limiting hydration with the cations bearing both an aryloxy group and an alkoxy group. 0
9
.
A ~ ~ - O C H , ArOH
showing a reaction with water and hydroxide ion. As illustrated in Figure 2, significant rate accelerations due to the base component of the buffer were also observed. Catalytic coefficients k , were determined as the slopes of plots of kobsdvs. base concentration. The constants ko, koH, and kA are listed in Tables I and 11. Rate constants were also measured for selected cations in sulfuric acid solutions and sodium perchlorate solutions and are listed in Tables I11 and IV.
Discussion By analogy with previous investigation^'^^^^^^ we assume that a pentavalent hydroxyphosphorane intermediate is involved in the hydrolysis. The exception of course is that portion of the water
I. I
PhP-
I.
PhP-
I
OR
I
Ph P'
I'
OR
It 0
II
PhP-
I
reaction which proceeds with C-0 cleavage. We also feel that formation of the intermediate is rate limiting in all cases, so that the observed rate constants refer to kl of eq 7 . The nature of the rate-limiting step is determined by the relative magnitudes of the rate constants kl and k2 for breakdown of the intermediate. The condition for the first step to be rate limiting is that k2 be greater than k-,. As has been argued previously for the hydrolysis of simple o x o c a r b o ~ a t i o n sbreakdown , ~ ~ ~ ~ ~ in the forward direction should be favored. In this direction the hydroxyl group remains on the phosphorus. Proton loss can be concerted with loss of the (35) Chiang, Y.; Kresge, A. J.; Lahti, M. 0.; Weeks, D. P. J . Am. Chem. SOC.1983, 105, 6852-6855.
~H~ A~PLOCH,
LA,
k,
e k.!
CH,
OH AA:OCH~ I CH, OAr
(7)
+k
f Ary-OAr
CH,OH
CH,
The products imply that the intermediate undergoes breakdown in the forward direction with only loss of the aryloxy group, so that kZA'of eq 7 is greater than kZCH3.Since similar leaving groups are involved, loss of methoxy and loss of hydroxy should occur a t similar rates, or in other words k-, is about the same as kZCH3. In consequence one can argue that kZAris also greater than k-, so that the intermediate loses only the aryloxy group and its formation is rate limiting. The lack of dependence on acidity of the products from the mixed cation 15 is interesting in light of the mechanism suggested by Kubisen and Westheimer (see eq 2),9 since if this were to apply to 15, strong acids such as 35% H2S04might be expected to produce some methanol and aryl ester. One explanation is that with the mixed phosphonium ion water addition produces a pentavalent intermediate with the phenol This is well set up for loss of the phenyl group, but a pseudorotation must occur for loss of methanol. The hydrolysis involving the base component of the buffer represents general base catalyzed addition of water to form the hydroxyphosphorane. An alternative mechanism involves nu-
/
A-
A
cleophilic catalysis where the buffer adds to the phosphonium ion. (36) Brierley, J.; Trippett, S.; White, M. W. J . Chem. SOC.,Perkin Trans.
I 1911, 213-278.
5208 J . A m . Chem. Soc., Vol. 107, No. 18, 1985
McClelland, McGall, and Patel
Table 11. Rate Constants for the Hydrolysis of the Arylmethyldimethoxyphosphonium Ions 8 (Ar = 4-CIC6H4)and 9 (Ar = 4-CH3C6H4)(25 OC, p = 0.5), the Phenyltrimethoxyphosphonium Ion (10) (25 OC, p = 0.5), the Diphenyldimethoxyphosphonium Ion ( 1 2 ) (25"C, p = O . l ) , the Diphenylmethylmethoxyphosphonium Ion (13) (25 "C, p = O.l), and the 4-Methoxyphenoxyphosphonium Ions 14 and 15 (25 "C, p = 2.0) k (M-I s-I) 8
H2O CNCH2C02ClCH2C02CH,OCH,CO; CHpC02(CHd2As02-
co
